Tool
Neo4j is a native graph database. In the cited discussion, it is used as an example of a graph-database implementation, which Mike Stonebraker characterizes as less performant than relational representations of graphs.
8 uses taken from transcripts — each links to the moment in the video.
A native graph database discussed as an example of graph-database implementations that Stonebraker considers less performant than relational representations of graphs.
Stores movie data, personal memories, exhibitors, notes, and relationships as graph data.
Entity graph database suggested for graph/relation memory with traversal-based retrieval
Graph database storing the agent context graph: short-term, long-term and reasoning memory with typed relationships and decision traces
Graph database storing code nodes and typed relationship edges, queried with Cypher
The graph database used in collaboration with Spranic to store the semantic representation of scanned buildings and connect observations across repeat visits.
Graph database platform presented as the basis for 'graph engineering': designing agent work as explicit graphs. Neo4j defines a knowledge layer that sits between systems of record (Databricks, BigQuery, Snowflake, docs) and agent runtimes, holding context graphs, agent memory, and graph-shaped skills so agents are predictable, governable and portable. Includes the core database plus AI/memory tooling that wraps around it.
Graph database platform whose knowledge-graph technology the talk uses to map an entire codebase as a graph, capturing relationships like batch jobs, copybooks and outputs so AI agents get accurate multi-hop context.
8 in the library.